OpenAI is rolling out parallel agents in ChatGPT to eligible Pro users. Multiple agents can work on different parts of a task, with progress visible in the conversation and completed tasks available for individual review. 36Kr reported that OpenAI announced the update on October 16, 2026. OpenAI’s release notes describe the feature and its rollout; 36Kr reported the announcement date.
How parallel agents work in ChatGPT
The feature is for GPT-6 Pro in ChatGPT’s Chat tab. Instead of keeping a task’s work in one response, multiple agents can work on different parts in parallel. You can follow their progress in the conversation, then open completed tasks to review what each agent did.
That gives you a way to inspect individual work as it finishes, rather than treating the result as a single, unexplained answer. OpenAI also describes the GPT-6 Pro update as bringing faster responses and improved use of plugins and tools.
Who can access the feature
OpenAI says the feature is rolling out to eligible Pro users. In managed workspaces, access depends on workspace settings as well as rollout eligibility. The announcement does not establish access for every Pro account.
OpenAI’s model guide identifies GPT-6 Pro as powered by GPT-6 Astra and lists the model for Pro $100, Pro $200, Business and Enterprise plans. Enterprise model access is subject to workspace permissions. Those plan details concern GPT-6 Pro; the feature’s rollout has its own eligibility terms. OpenAI’s model guide.
For background on the underlying model, see NeoTeo’s earlier coverage of GPT-6 Astra.
